On Apr 5, 2008, at 2:23 PM, Elwell, John wrote:
>>
>> On Apr 5, 2008, at 12:08 PM, Dean Willis posited:
>>>
>> Or we could fix it by either fundamentally changing RFC 4474, or by
>> providing a different fingerprint integrity mechanism for DTLS-SRTP
>> framework and thereby breaking the dependency cycle.
> [JRE] But isn't this effectively what the Wing and Fischer drafts are
> doing, i.e., providing a different fingerprint integrity mechanism?  
> They
> would probably still need something like a PSTN disclaimer solution  
> too.

It's a different thing to assert "this media goes with this  
INVITE" (when you sent both) than it is to assert "The identity we  
have authenticated as the source of this call is +19725199190" (when  
you don't really know).

The gateway service provider sending the INVITE is certainly  
responsible for the INVITE, so if they want to cryptographically  
assert that the RTP they're also responsible for is related to that  
INVITE, then I don't see that as a liability. Of course if they get it  
wrong they might have a problem, but that's just a cost of doing  
business -- and it's a lot easier to make assertions about something  
you control (the INVITE and the RTP) than about something you don't  
(the Caller-ID information).

In other words, if you're not asserting something as true that you  
reasonably have cause to expect may not be true, then you probably  
have no negligence, at least not of that type I've been talking about.

Do we need a special disclaimer for populating a From header field  
with a phone number? As this is the Best Current Practice, and since  
we've gone to some pains in the industry to say that an  
unauthenticated From:  is pretty much meaningless (see all the 3GPP  
sample calls from AlienBlaster), my opinion is no, we wouldn't need a  
disclaimer.

As for what the drafts actually do (do either Wing or Fischer  
completely eliminate a dependency on RFC 4474?), I'll leave that to  
the working group to figure out. Although it sure looks to me like  
that's what they're trying to do. :-).

--
Dean

_______________________________________________
Sip mailing list  https://www.ietf.org/mailman/listinfo/sip
This list is for NEW development of the core SIP Protocol
Use [EMAIL PROTECTED] for questions on current sip
Use [EMAIL PROTECTED] for new developments on the application of sip

Reply via email to